Servoss and Finn relied on a sample of about 700 high schools, drawing from the data of three national surveys.
A sample of about 36 in each group was found to be adequate to detect a 10% change in the Oswestry Low Back Pain Disability Questionnaire assuming a between patient variability of 15% [8], [9], a two-sided significance level of 5%, and a power of 80%.
The goal was to obtain a sample of about 2,000 interviews, i.e., slightly more than 1 per 30 GPs.
Following this survey, the UMCP has since been restructured accordingly, with habitat surveillance reduced to a sample of about 6% of TCUs each week.
Furthermore, it seems likely that any modification will be modest indeed if it cannot be detected in a sample of about 4,000 (ALSPAC).
A sample of about 0.5 cm was cut from the centre of the latest fully developed leaf of each plant, immediately frozen in liquid nitrogen, and stored at −80°C.
The frequently appearing statement in newspapers in the United States that a given opinion poll involving a sample of about 1,100 persons has a sampling error of no more than 3 percent is based on this kind of calculation.
Previous surveys relied on a sample of about 18,000 people who were questioned in person or through paper forms.
But this was based on a sample of about thirteen thousand people — a far cry from a national census.
The findings, based on a sample of about 25,000 inmates, were drawn from personal interviews and prisoners' own reports of symptoms, psychiatric treatments or medications.
Some less exact comparisons can be drawn from the writing tests, as a sample of about 1,500 schools had their test marked externally.
